k8s-cluster-ops
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides extensive capabilities to execute kubectl commands that can modify cluster state, including 'apply', 'delete', 'scale', and 'patch' operations across deployments, pods, and services.
- [CREDENTIALS_UNSAFE]: The skill facilitates access to the Kubernetes configuration file (kubeconfig), which contains sensitive authentication material. It also documents the use of 'kubectl get secrets' to retrieve sensitive data stored within the cluster.
- [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by instructing the agent to read and process untrusted data from the cluster, such as container logs ('kubectl logs') and event descriptions ('kubectl describe'), which could be manipulated by an attacker with cluster access.
- [COMMAND_EXECUTION]: Includes helper scripts ('scripts/') that automate complex workflows like node draining and pod debugging, involving multiple sequential kubectl calls.
Audit Metadata